gpio/omap v2: map irq_enable/disable to mask/unmask.

gpio/omap v2: map irq_enable/disable to mask/unmask.

[Andreas Fenkart]

Hi Santosh,

I submitted the following patch a while back.
https://patchwork.kernel.org/patch/1886421/

As already said, the patch is straight forward, but without it,
we probably will not see decent SDIO performance on am335x chips.

[why it is needed]

The omap_hsmmc module is suspended whenever it is idle, its
functional clock being turned off. In this mode it is not able to
forwared IRQs to the system. For that to happen, it needs to tell
the PRCM to restore it's fclk.

                   ------
                  | PRCM |
                   ------
                    | ^
               fclk | | swakeup
                    v |
                  -------               ------
      <-- IRQ -- | hsmmc | <-- CIRQ -- | card |
                  -------               ------

This is done through the swakeup line, which can be configured to
trigger for various events, among others; CIRQ. The problem is
that on the AM335x family the swakeup line is missing, it has not
been routed from the module to the PRCM.

[solution]
the simplest solution was to keep the fclk enabled all the
time. But that was not accepted, instead this was suggested

> > The alternative was to configure dat1 line as a GPIO, while
> > waiting for an IRQ. Then configuring it back as dat1 to serve
> > the SDIO card after it signalled an IRQ. Or when the host
> > wants to start a transfer.
>
> The way to implement this is set named states in the .dts file
> for the pins using pinctrl-single.c, then have the MMC driver
> request states "default" "active" and "idle" during the probe,
> then toggle between active and idle during the runtime.

Surprisingly the induced overhead is quite small, the performance
is similar to keeping the fclk enabled at all times. See here
for full thread:
http://www.spinics.net/lists/linux-omap/msg83363.html
https://patchwork.kernel.org/patch/1901471/  ... or just the patch

There are still open questions to gpio patch itself, see here
http://www.mail-archive.com/linux-omap@vger.kernel.org/msg87217.html

So could we pls reopen that thread? It might be on the other side
of your mailbox

rgds,
Andi

Re: gpio/omap v2: map irq_enable/disable to mask/unmask.

[Tony Lindgren]

Hi,

* Andreas Fenkart <andreas.fenkart@streamunlimited.com> [130411 06:23]:
> Hi Santosh,
> 
> I submitted the following patch a while back.
> https://patchwork.kernel.org/patch/1886421/
> 
> As already said, the patch is straight forward, but without it,
> we probably will not see decent SDIO performance on am335x chips.

I suggest you repost, patches can easily get forgotten unuless
you follow-up getting them merged.

> [why it is needed]
> 
> The omap_hsmmc module is suspended whenever it is idle, its
> functional clock being turned off. In this mode it is not able to
> forwared IRQs to the system. For that to happen, it needs to tell
> the PRCM to restore it's fclk.
> 
>                    ------
>                   | PRCM |
>                    ------
>                     | ^
>                fclk | | swakeup
>                     v |
>                   -------               ------
>       <-- IRQ -- | hsmmc | <-- CIRQ -- | card |
>                   -------               ------
> 
> This is done through the swakeup line, which can be configured to
> trigger for various events, among others; CIRQ. The problem is
> that on the AM335x family the swakeup line is missing, it has not
> been routed from the module to the PRCM.
> 
> [solution]
> the simplest solution was to keep the fclk enabled all the
> time. But that was not accepted, instead this was suggested
> 
> > > The alternative was to configure dat1 line as a GPIO, while
> > > waiting for an IRQ. Then configuring it back as dat1 to serve
> > > the SDIO card after it signalled an IRQ. Or when the host
> > > wants to start a transfer.
> >
> > The way to implement this is set named states in the .dts file
> > for the pins using pinctrl-single.c, then have the MMC driver
> > request states "default" "active" and "idle" during the probe,
> > then toggle between active and idle during the runtime.
> 
> Surprisingly the induced overhead is quite small, the performance
> is similar to keeping the fclk enabled at all times. See here
> for full thread:
> http://www.spinics.net/lists/linux-omap/msg83363.html
> https://patchwork.kernel.org/patch/1901471/  ... or just the patch

That's nice, AFAIK this is the only way to do it for some omaps.
 
> There are still open questions to gpio patch itself, see here
> http://www.mail-archive.com/linux-omap@vger.kernel.org/msg87217.html
> 
> So could we pls reopen that thread? It might be on the other side
> of your mailbox

The SDIO related patch can already get merged while the GPIO patch
is being discussed I hope?

If so, you should fix #if 0 part I commented on and repost with
the ack from Grant. And update it against the current linux next
so people can apply it for testing.

Regards,

Tony

Re: gpio/omap v2: map irq_enable/disable to mask/unmask.

[Santosh Shilimkar]

On Thursday 11 April 2013 06:48 PM, Andreas Fenkart wrote:
> Hi Santosh,
> 
> I submitted the following patch a while back.
> https://patchwork.kernel.org/patch/1886421/
> 
> As already said, the patch is straight forward, but without it,
> we probably will not see decent SDIO performance on am335x chips.
> 
> [why it is needed]
> 
Why part is clear for me.

> There are still open questions to gpio patch itself, see here
> http://www.mail-archive.com/linux-omap@vger.kernel.org/msg87217.html
> 
> So could we pls reopen that thread? It might be on the other side
> of your mailbox
> 
Your idea is reasonable and probably patch make lot of sense for
edge triggered interrupt.

Can you please repost the patch with the more comprehensive change-log
as discussed in the thread.

Regards,
Santosh

Re: gpio/omap v2: map irq_enable/disable to mask/unmask.

[Jon Hunter]

On 04/11/2013 10:54 AM, Santosh Shilimkar wrote:
> On Thursday 11 April 2013 06:48 PM, Andreas Fenkart wrote:
>> Hi Santosh,
>>
>> I submitted the following patch a while back.
>> https://patchwork.kernel.org/patch/1886421/
>>
>> As already said, the patch is straight forward, but without it,
>> we probably will not see decent SDIO performance on am335x chips.
>>
>> [why it is needed]
>>
> Why part is clear for me.
> 
>> There are still open questions to gpio patch itself, see here
>> http://www.mail-archive.com/linux-omap@vger.kernel.org/msg87217.html
>>
>> So could we pls reopen that thread? It might be on the other side
>> of your mailbox
>>
> Your idea is reasonable and probably patch make lot of sense for
> edge triggered interrupt.
> 
> Can you please repost the patch with the more comprehensive change-log
> as discussed in the thread.

Originally, I had a question on enable/disable versus mask/unmask, but
looking at the way we have implemented mask/unmask for omap, I don't
think that my question is really applicable. Yes so please re-post.

Jon